scholarly journals Applied Comparison of Meta-analysis Techniques

10.36469/9848 ◽  
2013 ◽  
Vol 1 (1) ◽  
pp. 14-22
Author(s):  
Li Wang ◽  
Colin Lewis-Beck ◽  
Elyse Fritschel ◽  
Erdem Baser ◽  
Onur Baser
Keyword(s):  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Fixed Effects Model ◽  
Analysis Techniques ◽  
Level Data ◽  
Study Results ◽  
Meta Regression ◽  
Treatment Comparisons ◽  
Analysis Models

Background: Meta-analysis is an approach that combines findings from similar studies. The aggregation of study level data can provide precise estimates for outcomes of interest, allow for unique treatment comparisons, and explain the differences arising from conflicting study results. Proper meta-analysis includes five basic steps: identify relevant studies; extract summary data from each paper; compute study effect sizes, perform statistical analysis; and interpret and report the results. Objectives: This study aims to review meta-analysis methods and their assumptions, apply various meta-techniques to empirical data, and compare the results from each method. Methods: Three different meta-analysis techniques were applied to a dataset looking at the effects of the bacille Calmette-Guerin (BCG) vaccine on tuberculosis (TB). First, a fixed-effects model was applied; then a random-effects model; and third meta-regression with study-level covariates were added to the model. Overall and stratified results, by geographic latitude were reported. Results: All three techniques showed a statistically significant effects from the vaccination. However, once covariates were added, efficacy diminished. Independent variables, such as the latitude of the location in which the study was performed, appeared to be partially driving the results. Conclusions: Meta-analysis is useful for drawing general conclusions from a variety of studies. However, proper study and model selection are important to ensure the correct interpretation of results. Basic meta-analysis models are fixed-effects, random-effects, and meta-regression.

The relationship between migraine and infant colic: A systematic review and meta-analysis

Cephalalgia ◽  
2014 ◽  
Vol 35 (1) ◽  
pp. 63-72 ◽  
Author(s):  
Amy A Gelfand ◽  
Peter J Goadsby ◽  
I Elaine Allen
Keyword(s):  
Systematic Review ◽  
Random Effects ◽  
Odds Ratio ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Primary Analysis ◽  
Infant Colic ◽  
The Relationship

Context Infant colic is a common and distressing disorder of early infancy. Its etiology is unknown, making treatment challenging. Several articles have suggested a link to migraine. Objective The objective of this article was to perform a systematic review and, if appropriate, a meta-analysis of the studies on the relationship between infant colic and migraine. Data sources Studies were identified by searching PubMed and ScienceDirect and by hand-searching references and conference proceedings. Study selection For the primary analysis, studies specifically designed to measure the association between colic and migraine were included. For the secondary analysis, studies that collected data on colic and migraine but were designed for another primary research question were also included. Data extraction Data were abstracted from the original studies, through communication with study authors, or both. Two authors independently abstracted data. Main outcomes and measures The main outcome measure was the association between infant colic and migraine using both a fixed-effects model and a more conservative random-effects model. Results Three studies were included in the primary analysis; the odds ratio for the association between migraine and infant colic was 6.5 (4.6–8.9, p < 0.001) for the fixed-effects model and 5.6 (3.3–9.5, p = 0.004) for the random-effects model. In a sensitivity analysis wherein the study with the largest effect size was removed, the odds ratio was 3.6 (95% CI 1.7–7.6, p = 0.001) for both the fixed-effects model and random-effects model. Conclusions In this meta-analysis, infant colic was associated with increased odds of migraine. If infant colic is a migrainous disorder, this would have important implications for treatment. The main limitation of this meta-analysis was the relatively small number of studies included.

Meta-Analysis of Meta-Analyses in Communication: Comparing Fixed Effects and Random Effects Analysis Models

2010 ◽  
Vol 58 (3) ◽  
pp. 257-278 ◽  
Author(s):  
Ashley Anker ◽  
Amber Marie Reinhart ◽  
Thomas Hugh Feeley
Keyword(s):  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Meta Analyses ◽  
Analysis Models

scholarly journals A review and meta-analysis of red and processed meat consumption and breast cancer

2010 ◽  
Vol 23 (2) ◽  
pp. 349-365 ◽  
Author(s):  
Dominik D. Alexander ◽  
Libby M. Morimoto ◽  
Pamela J. Mink ◽  
Colleen A. Cushing
Keyword(s):  
Breast Cancer ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Random Effect ◽  
Red Meat ◽  
Processed Meat ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Meat Intake

The relationship between meat consumption and breast cancer has been the focus of several epidemiological investigations, yet there has been no clear scientific consensus as to whether red or processed meat intake increases the risk of breast cancer. We conducted a comprehensive meta-analysis incorporating data from several recently published prospective studies of red or processed meat intake and breast cancer. In the meta-analysis utilising data from the Pooling Project publication (includes data from eight cohorts) combined with data from nine studies published between 2004 and 2009 and one study published in 1996, the fixed-effect summary relative risk estimate (SRRE) for red meat intake (high v. low) and breast cancer was 1·02 (95 % CI 0·98, 1·07; P value for heterogeneity = 0·001) and the random-effects SRRE was 1·07 (95 % CI 0·98, 1·17). The SRRE for each 100 g increment of red meat was 1·04 (95 % CI 1·00, 1·07), based on a fixed-effects model, and 1·12 (95 % CI 1·03, 1·23) based on a random-effects model. No association was observed for each 100 g increment of red meat among premenopausal women (SRRE 1·01; 95 % CI 0·92, 1·11) but a statistically significant SRRE of 1·22 (95 % CI 1·04, 1·44) was observed among postmenopausal women using a random-effects model. However, the association for postmenopausal women was attenuated and non-significant when using a fixed-effects model (SRRE 1·03; 95 % CI 0·98, 1·08). The fixed- and random-effect SRRE for high (v. low) processed meat intake and breast cancer were 1·00 (95 % CI 0·98, 1·01; P value for heterogeneity = 0·005) and 1·08 (95 % CI 1·01, 1·16), respectively. The fixed- and random-effect SRRE for each 30 g increment of processed meat were 1·03 (95 % CI 1·00, 1·06) and 1·06 (95 % CI 0·99, 1·14), respectively. Overall, weak positive summary associations were observed across all meta-analysis models, with the majority being non-statistically significant. Heterogeneity was evident in most analyses, summary associations were sensitive to the choice of analytical model (fixed v. random effects), and publication bias appeared to have produced slightly elevated summary associations. On the basis of this quantitative assessment, red meat and processed meat intake does not appear to be independently associated with increasing the risk of breast cancer, although further investigations of potential effect modifiers, such as analyses by hormone receptor status, may provide valuable insight to potential patterns of associations.

Abstract 266: Quantifying the Effect of Technological Advancements and Guidelines on Pediatric Extracorporeal Cardiopulmonary Resuscitation Survival Outcomes

Circulation ◽  
2020 ◽  
Vol 142 (Suppl_4) ◽  
Author(s):  
Anish Sangari ◽  
Nitish Sood ◽  
Joseph Hauger
Keyword(s):  
Regression Analysis ◽  
Cardiopulmonary Resuscitation ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Survival Rates ◽  
Serum Lactate ◽  
Fixed Effects Model ◽  
Extracorporeal Cardiopulmonary Resuscitation ◽  
Meta Regression ◽  
Meta Regression Analysis

Introduction: The 2009 H1N1 pandemic resulted in a substantial increase in the use of extracorporeal cardiopulmonary resuscitation (ECPR). The concurrent release of guidelines by the French Ministry of Health also led to the standardization of ECPR protocol. We conducted a systematic review and meta-analysis to examine if the simultaneous development of guidelines and expanded usage led to an increase in survival to hospital discharge (SHD) in pediatric ECPR and to investigate if predictors of survival have changed since 2009. Methods: PubMed, SCOPUS, Google Scholar, Cochrane, and EMBASE were searched to identify studies examining survival outcome and predictors of survival for pediatric ECPR. A random-effects model was applied. Additional pre-specified subgroup analysis was performed using a fixed-effects model. A univariate meta-regression analysis was conducted. Results: The meta-analysis included 3,454 patients from 30 studies. Total survival rates have not improved in pediatric ECPR between subgroups (Mann-Whitney U test: p > 0.05). Extracorporeal membrane oxygenation (ECMO) duration for survivors was not significantly different than nonsurvivors before 2009 RR -0.02 [95% CI: -0.19 - 0.14], but was significantly lower than nonsurvivors after 2009 RR -0.32 [-0.48 - -0.15, p < 0.05]. Meta-regression analysis revealed the ratio of ECMO duration for survivors compared to non-survivors has significantly decreased from 1998 to 2020 (p < 0.05). Patient age was a significant predictor of survival pre-2009 RR but is not correlated with survival post-2009 (p > 0.05). PaO2 was significantly higher in survivors than nonsurvivors pre-2009 RR 0.15 [0.12 - 0.18, p < 0.05] but was not significantly different post-2009 (p > 0.05). End-tidal CO2, CPR duration, pre-CPR serum lactate, and pre-CPR creatinine were not significantly different (p > 0.05) between survivors and non-survivors in either pre-2009 or post-2009. Conclusion: Pediatric ECPR did not see major improvement in survival rates since 2009. In addition, predictors of survival for pediatric ECPR cases have changed over the last 10 years. As such, clinicians may be advised to rely much more heavily on recent research regarding predictors of survival in pediatric ECPR cases.

scholarly journals Comparison of coated meshes for intraperitoneal placement in animal studies: a systematic review and meta-analysis

Hernia ◽  
2019 ◽  
Vol 24 (6) ◽  
pp. 1253-1261 ◽  
Author(s):  
H. Liu ◽  
S. van Steensel ◽  
M. Gielen ◽  
T. Vercoulen ◽  
J. Melenhorst ◽  
...  
Keyword(s):  
Systematic Review ◽  
Random Effects ◽  
Fixed Effects ◽  
Animal Studies ◽  
Meta Analysis ◽  
Adhesion Formation ◽  
Mean Difference ◽  
Random Effects Model ◽  
Fixed Effects Model

Abstract Purpose Laparoscopic intraperitoneal onlay mesh in hernia repair can result in adhesions leading to intestinal obstruction and fistulation. The aim of this systematic review is to compare the effects of mesh coatings reducing the tissue-to-mesh adhesion in animal studies. Methods Pubmed and Embase were systematically searched. Animal experiments comparing intraperitoneally placed meshes with coatings were eligible for inclusion. Only studies with comparable follow-up, measurements, and species were included for data pooling and subsequent meta-analysis. Results A total of 131 articles met inclusion criteria, with four studies integrated into one comparison and five studies integrated into another comparison. Compared to uncoated polypropylene (PP) mesh, PP mesh coated with hyaluronic acid/carboxymethyl cellulose (HA/CMC) showed significantly reduced adhesion formation at follow-up of 4 weeks measured with adhesion score of extent (random effects model, mean difference,−  0.96, 95% CI − 1.32 to − 0.61, P < 0.001, I2 = 23%; fixed effects model, mean difference,− 0.94, 95% CI − 1.25 to − 0.63, P < 0.001, I2 = 23%). Compared to PP mesh, polyester mesh coated with collagen (PC mesh) showed no significant difference at follow-up of 4 weeks regarding percentage of adhesion-area on a mesh, using random effects model (mean difference − 11.69, 95% CI − 44.14 to 20.76, P = 0.48, I2 = 92%). However, this result differed using fixed effects model (mean difference − 25.55, 95% CI − 33.70 to − 7.40, P < 0.001, I2 = 92%). Conclusion HA/CMC coating reduces adhesion formation to PP mesh effectively at a follow-up of 4 weeks, while the anti-adhesive properties of PC mesh are inclusive comparing all study data.

scholarly journals Association between Genetic Mutations and Risk of Venous Thromboembolism in Patients with Solid Tumor Malignancies: A Systematic Review and Meta-Analysis

Blood ◽  
2020 ◽  
Vol 136 (Supplement 1) ◽  
pp. 33-34
Author(s):  
Mohammed Abufarhaneh ◽  
Rudra Kashyap Pandya ◽  
Ahmed Alkhaja ◽  
Alla Iansavitchene ◽  
Stephen Welch ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Venous Thromboembolism ◽  
Relative Risk ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Genetic Mutations ◽  
Fixed Effects Model ◽  
Wild Type ◽  
Small Cell Lung

Title: Association between genetic mutations and risk of venous thromboembolism in patients with solid tumor malignancies: A systematic review and meta-analysis. Introduction: Venous thromboembolisms (VTE) is a frequent complication in cancer patients with an overall incidence of approximately 20% and is associated with significant morbidity, mortality, and burden on the health care system [1]. VTE risk varies according to tumor type and it is possible that certain genetic mutations may promote hypercoagulability resulting in a prothrombotic state. We have studied the association between genetic mutations and risk of VTE in patients with melanoma, small cell lung cancer (SCLC), non-small cell lung cancer (NSCLC), colon, gastric and ovarian cancers. Aim: To evaluate the association of genetic mutations in melanoma, lung (SCLC, NSCLC), colon, gastric and ovarian cancer with the risk of venous thromboembolism. Search strategy: We conducted a systematic literature search in MEDLINE and EMBASE electronic databases from inception to May 2020 using a combination of the following subject headings and text words: gene mutation, genetic polymorphism, oncogenes, tumor-related genes, proto-oncogenes, oncoproteins, venous thromboembolism, venous thrombosis, pulmonary embolism, lung neoplasms (SCLC, NSCLC), gastrointestinal neoplasms, ovarian neoplasms, colorectal neoplasms, and melanoma. Selection criteria: We included studies presenting data on genetic mutations with &gt; 5% prevalence in our population of interest [adult patients with melanoma, lung (SCLC, NSCLC), colon, gastric and ovarian cancer]. Studies with thrombosis events given, but not listed according to the mutation status of patients were excluded. Retrospective and prospective cohort studies, case-control studies and randomized clinical trials (RCTs) were included. Data collection and analysis: General descriptors for the patients in each study were collected, as well as number of patients with wildtype or mutated genes of interest that developed thrombosis. Fixed and random effects models were used to generate pooled adjusted ratios with 95% confidence intervals using data for thrombosis events in patients with mutated and wildtype driver genes of interest. The quality of the studies: Study qualities were assessed using Jadad score for randomized controlled trials and the Newcastle-Ottawa Scale for case-control and cohort studies. Main results: Of 616 eligible articles, we included 31 articles based on our inclusion criteria and 10 were included in the meta-analysis. In patients with lung cancer with EGFR, KRAS and ALK mutations the relative risk (RR) of VTE was 1.006 (40.084-11.150, P=0.965), 1.034 (0.514-2.081, P=0.925) and 1.571 (1.314-1.878, P&lt;0.001) respectively using the fixed-effects model. Using the random-effects model, the RR of VTE in patients with lung cancer and EGFR, KRAS and ALK was 1.058 (0.632-1.768, P=0.831), 1.086 (0.504-2.343, P=0.833) and 1.667 (1.303-2.131, P&lt;0.001), respectively. In patients with colon cancer and KRAS mutation the RR of VTE was 1.38 (1.368-33.942, P=0.017) using a fixed-effects model and 1.319 (0.794-2.191, P=0.285) using a random-effects model. Conclusion: In patients with lung cancer, those bearing ALK rearrangements carried a significantly higher relative risk of developing thrombosis than those with wild-type ALK. In patients with colon cancer, relative risk of VTE increased significantly in patients who had KRAS mutations than KRAS wild type. Disclosures No relevant conflicts of interest to declare.

scholarly journals The relationship between time to surgery after neoadjuvant chemotherapy and survival in breast cancer patients:a meta-analysis

2020 ◽  
Author(s):  
YongCheng Su ◽  
XiaoGang Zheng
Keyword(s):  
Neoadjuvant Chemotherapy ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Disease Free Survival ◽  
Significant Heterogeneity ◽  
Fixed Effects Model ◽  
Free Survival ◽  
The Impact ◽  
The Relationship

Abstract PURPOSE: This meta-analysis aims to evaluate the impact of delaying surgery in operable breast tumor patients after neoadjuvant chemotherapy (NAC) on survival. METHODS:An electronic literature retrieval was conducted on PubMed/Medline and EMBASE((between January 2000 and June 2020).The primary end point was overall survival(OS),secondary end points included disease-free survival (DFS) or recurence-free survival (RFS).The HR with 95% confidence intervals were calculated using a random-effects or fixed-effects model. RESULTS:The combined HR for OS was 1.51 (95% CI:1.30-1.76; P=0.000) by fixed effects model.No statistically significant heterogeneity was found (P=0.168, I 2 =31.3%).The pooled HR for RFS/DFS was 1.59 (95%CI:1.30-1.95,I 2 = 66.0%) by random-effects model,with significant heterogeneity. CONCLUSION:Our meta-analysis revealed a significant adverse association between longer TTS after NAC and more inferior OS and RFS/DFS in patients with breast cancer.Clinicians and patients should minimize surgical delay after NAC as much as possible.

scholarly journals Comparison of efficacy of robotic surgery, laparoscopy, and laparotomy in the treatment of ovarian cancer: a meta-analysis

2019 ◽  
Vol 17 (1) ◽  
Author(s):  
Can Shi ◽  
Yingchun Gao ◽  
Yijun Yang ◽  
Lei Zhang ◽  
Juanpeng Yu ◽  
...  
Keyword(s):  
Ovarian Cancer ◽  
Robotic Surgery ◽  
Publication Bias ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Cochrane Library ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Estimated Blood Loss

Abstract Background We intended to compare the clinical effect of robotic surgery with laparoscopy and laparotomy in ovarian cancer treatment. Methods The included studies were retrieved from PubMed, Embase, and the Cochrane Library databases. The Methodological Index for Nonrandomized Studies (MINORS) was used to evaluate the study quality. Effect measures were presented with weighted mean difference (WMD)/odds ratio (OR) and 95% confidence interval (CI), and heterogeneity test was assessed using Q test and I2 statistics to determine the use of the random effects model or fixed effects model. Egger’s test was used to assess the publication bias. Results A total of eight studies was included in this meta-analysis with a MINORS score of 16–18. In the random effects model, estimated blood loss (EBL) of robotic surgery was significantly less compared with laparotomy (WMD = − 521.7027, 95% CI − 809.7816; − 233.6238). In the fixed effects model, length of hospital stay (LHS) (WMD = − 5.2225, 95% CI − 6.1485; − 4.2965) and postoperative complication (PC) (OR = 0.4710, 95% CI 0.2537; 0.8747) of robotic surgery were significantly less, and overall survival (OS) rate (OR = 6.4355, 95% CI 1.6722; 24.7678, P = 0.0070) of robotic surgery was significantly higher compared with laparotomy. There was no difference in the effect size of all variables between robotic surgery and laparoscopy. Meanwhile, a publication bias (t = 6.8290, P = 0.002405) was only identified for PC in robotic surgery and laparotomy groups; no publication bias was identified for the other variables. Conclusions Despite the above results, it failed to show oncological safety and recurrence by pathological stages or histologic types in this meta-analysis, and those confounding factors might affect the clinical outcome. Future meta-analyses with a larger number of eligible randomized controlled trial studies were needed to determine the most suitable treatment method for patients with different stages and types of ovarian cancer.

Sign in / Sign up

Export Citation Format

Share Document